On many modern Web platforms users can annotate the available online resources with freely-chosen tags. This Social Tagging data can then be used for information organization or retrieval purposes. Tag recommenders in that context are designed to help the online user in the tagging process and suggest appropriate tags for resources with the purpose to increase the tagging quality. In recent years, different algorithms have been proposed to generate tag recommendations given the ternary relationships between users, resources, and tags. Many of these algorithms however suffer from scalability and performance problems, including the popular FolkRank algorithm. In this work, we propose a neighborhood-based tag recommendation algorithm called LocalRank, which in contrast to previous graph-based algorithms only considers a small part of the user-resource-tag graph. An analysis of the algorithm on a popular social bookmarking data set reveals that the recommendation accuracy is on a par with or slightly better than FolkRank while at the same time recommendations can be generated instantaneously using a compact in-memory representation.
Description
LocalRank - Neighborhood-Based, Fast Computation of Tag Recommendations | SpringerLink
%0 Book Section
%1 Kubatz2011
%A Kubatz, Marius
%A Gedikli, Fatih
%A Jannach, Dietmar
%B E-Commerce and Web Technologies: 12th International Conference, EC-Web 2011, Toulouse, France, August 30 - September 1, 2011. Proceedings
%C Berlin, Heidelberg
%D 2011
%E Huemer, Christian
%E Setzer, Thomas
%I Springer Berlin Heidelberg
%K ranking recommender
%P 258--269
%R 10.1007/978-3-642-23014-1_22
%T LocalRank - Neighborhood-Based, Fast Computation of Tag Recommendations
%U https://doi.org/10.1007/978-3-642-23014-1_22
%X On many modern Web platforms users can annotate the available online resources with freely-chosen tags. This Social Tagging data can then be used for information organization or retrieval purposes. Tag recommenders in that context are designed to help the online user in the tagging process and suggest appropriate tags for resources with the purpose to increase the tagging quality. In recent years, different algorithms have been proposed to generate tag recommendations given the ternary relationships between users, resources, and tags. Many of these algorithms however suffer from scalability and performance problems, including the popular FolkRank algorithm. In this work, we propose a neighborhood-based tag recommendation algorithm called LocalRank, which in contrast to previous graph-based algorithms only considers a small part of the user-resource-tag graph. An analysis of the algorithm on a popular social bookmarking data set reveals that the recommendation accuracy is on a par with or slightly better than FolkRank while at the same time recommendations can be generated instantaneously using a compact in-memory representation.
%@ 978-3-642-23014-1
@inbook{Kubatz2011,
abstract = {On many modern Web platforms users can annotate the available online resources with freely-chosen tags. This Social Tagging data can then be used for information organization or retrieval purposes. Tag recommenders in that context are designed to help the online user in the tagging process and suggest appropriate tags for resources with the purpose to increase the tagging quality. In recent years, different algorithms have been proposed to generate tag recommendations given the ternary relationships between users, resources, and tags. Many of these algorithms however suffer from scalability and performance problems, including the popular FolkRank algorithm. In this work, we propose a neighborhood-based tag recommendation algorithm called LocalRank, which in contrast to previous graph-based algorithms only considers a small part of the user-resource-tag graph. An analysis of the algorithm on a popular social bookmarking data set reveals that the recommendation accuracy is on a par with or slightly better than FolkRank while at the same time recommendations can be generated instantaneously using a compact in-memory representation.},
added-at = {2017-12-18T21:22:56.000+0100},
address = {Berlin, Heidelberg},
author = {Kubatz, Marius and Gedikli, Fatih and Jannach, Dietmar},
biburl = {https://www.bibsonomy.org/bibtex/220ff6cfd0555e2782ebfc2a2a1017bf8/hotho},
booktitle = {E-Commerce and Web Technologies: 12th International Conference, EC-Web 2011, Toulouse, France, August 30 - September 1, 2011. Proceedings},
description = {LocalRank - Neighborhood-Based, Fast Computation of Tag Recommendations | SpringerLink},
doi = {10.1007/978-3-642-23014-1_22},
editor = {Huemer, Christian and Setzer, Thomas},
interhash = {19a8194d47a5f6722a563a3689606440},
intrahash = {20ff6cfd0555e2782ebfc2a2a1017bf8},
isbn = {978-3-642-23014-1},
keywords = {ranking recommender},
pages = {258--269},
publisher = {Springer Berlin Heidelberg},
timestamp = {2017-12-18T21:22:56.000+0100},
title = {LocalRank - Neighborhood-Based, Fast Computation of Tag Recommendations},
url = {https://doi.org/10.1007/978-3-642-23014-1_22},
year = 2011
}